The movements of the wing in flight--Marey's experiments--Stopping and turning movements--Alighting--"Taking off"--Hovering--The use of the tail in flight--The carriage of the neck in flight--And of the legs--The flight of petrels--The speed of flight--The height at which birds fly--Flight with burdens--Experiments on the sizes of the wing in relation to flight--Flight in "troops." While it is possible to show that certain kinds of flight are to be associated with such and such peculiarities of the skeleton, and the muscles attached thereto, there are many "eccentricities" which cannot be measured, and explained, in terms of mechanism. The very disconcerting, twisting, flight of the snipe is one of these. The sportsman knows it well: and he knows that the twisting, during which the bird turns the body half over--that is with, say, the left wing pointing directly downwards, and the right wing directly upwards--is only the preliminary to getting fully on the way, and that, presently, it will pursue a straight course, with arrow-like speed. Yet its cousin, the jack-snipe, never twists. Why does the woodcock invariably drop after a charge of shot, even though not a pellet has touched it, while a snipe pursues its way? These differences are not merely differences of "habit": they indicate subtle differences in nervous response to the same kind of stimulus, and in structural details yet to be unravelled. Some day the cinematograph will reveal to us all the phases of flight and the movements to which they are due. Even now, thanks to the modern camera, we have learned a great deal. We have learned, for example, that the flight of a bird is not effected merely by rapid up and down movements of the fully extended wings, or with flexed wings--that is to say, half closed, as in "gliding" flight when a bird is descending, or in the swoop of, say, the sparrow-hawk. Only in one of these two positions do we ever seem to see the wings when we have to trust to our eyes alone, as the bird hurries past us. The impression that we have seen aright is confirmed when we stand on the deck of a steamer, and watch the gulls following in its wake. For incredibly long distances they will travel without a perceptible wing-beat. The albatross is the finest of all performers in regard to this kind of flight, which is due, apparently, to air currents created by stiff breezes, or gales. Some birds seem to make their way against a head-wind with the minimum of effort, by partly flexing the wings and gliding downwards: at the end of the descent, by turning the body sharply upwards, and spreading the wings to the fullest extent, they are lifted up, and driven forward, like a kite. Marey and Pettigrew, long ago, showed conclusively, by means of photography, that our conception of the movement of the wing during flight was far from correct.